CS525 S2000 Midterm Solution

Enter your answers on the web page and after completing your answers, push the submit button.  You have until 3/18/2000 4:30pm to finish the midterm.Treat these questions as mulitple-choice questions. You must choose either yes or no for each answer. There are two questions that require you to enter answers to the textareas.
Print a copy of your answers before hitting the submit button.   If you have problem accessing the web server for submitting the answer, slide your hard copy with answers under my office door.
Make sure the last line of the  return web page indicating your login is correct.  If not,  click back button to return to your midterm web page and retype your login/password.   If you still have problem, submit hard copy of your answer.
The login and password are those of the CS Unix machines, not those of brain.uccs.edu or mail.uccs.edu.
Enter the following information. The password is used to verify the person submitting the answers.
Your name:
Your login on CS UNIX machines:
Your password on CS UNIX machines:
Work/Home Phone number (option):

  1. Introduction to Video
    1. What is NTSC?
      2. 1. a video signal standard used in Japan Yes No
      2. 6MHz is allocated per channel Yes No
      3. video and audio signals are sent at different frequencies Yes No
      4. Each field contains 525 scanning line
      Each frame has 525 scanning line. and two fields.      		 Yes No
    2. Relationship among bandwidth, scaning lines, frame rates, horizontal blanking (retract) period, horizontal resolution.

      3. 1. If other parameters are fixed,  more scanning lines implies more video bandwidth needed  Yes No
      2. If other parameters are fixed,  more scanning lines implies fewer frames per second Yes No
      3. If other parameters are fixed,  longer horizontal blanking period implies lower frame rates Yes No
      4. If other parameters are fixed,  longer horizontal blanking period implies less resolution. Yes No
    3. TV vs. 35mm film

      4. 1. TV has higher resolutions Yes No
      2. TV has higher aperture response on wider (thick) TV line patterns. Yes No
      3. TV has higher aperture response on narrower (thin) TV line patterns. Yes No
    4. The Tungsen  light consists of lights of many colors.  Yes No
    5. Satuation of color in a light source indciates how pure is its color? Yes No
    6. YIQ video signal
      7. 1. Human eye are more sensitive to the luminence than color satuation Yes No
      2. Human eye are more sensitive to I signal than Q signal. Yes No
      3. When digitized TV signal, it makes sense to give more bits to Q signal than Y signal.  Yes No
      4. These three signal components are mixed during TV broadcast transmission. Yes No
  2. JPEG
    1. Zig-Zag Sequence and Run length Coding.

      2.         1.What is the intermediate encoding sequence for the computed DCT coefficients with 128, followed by 62 zeros and -17 ?


              2. How many  bits will be used  to encode this intermediate encoding sequence?

              3. Consider that 8 bits are used to encode the value of each original 64 pixels. What is compression ratio achieved by the
                combination of DCT, runlength and VLI coding?
  3. MPEG
    1. MPEG encoded frames generate fixed number of  bits per second.  Yes No
    2. Which of the following approaches can always achieve  higher compression ratio?
        3.  
        1. Reduce the number of quantization steps in the encoding of I frames Yes No
        2. Increase the number  of B frames in the encoding  frame sequence
        Depends on the content of the video.  Proper choices of reference frames for B frames
        criticial.  When the reference frame(s) is separated too far from the encoded frame we will
        not find macroblock match that above threshold.  The themes are just different.        		 Yes No
        3. Use motion compensation results of Y signal frames for  those of I and Q signal frames,
            i.e., no need to encode motion vectors for I and Q signal frames        		 Yes No
    3. Using homework 3 part 1.2 as an example, analyze the impact on the compression ratio by the above approaches. For each chosen approach, show the compression ratio improvement that can be achieved.


      4. 1. By reducing the number of quantization steps, we effectively use fewer bits to encode the DCT coefficients. If   16 quantization steps  is used instead of 256 quantization steps, the # of bits to encode the coefficients is reduced to just 4 bits. For hw3 part1.2,  the DC coefficient value is already zero and the VLI encoding has the effect of using fewer bits for smaller values.  Therefore reducing the number of quantization steps has no effect on the compression ratio in this case.

      2. Depends on the content of the video.  Proper choices of reference frames for B frames criticial.  When the reference frame(s) is separated too far from the encoded frame we will not find macroblock match that above threshold.  The themes are just different. When there is no match, we need to use intraframe coding.

      3. If I and Q signal frames are encoded separately, then we will have 2 additional <macro type><motion vector>[<motion vector>]<prediction error matrix> bit sequences.   By using Y signal's macroblock matching result, we effective using 35 bits to encode 3*(8*16*16)=6144 bits of data in 3 signal video frame sequences.  The improved compression  ratio is 6144/35=175.54.
       

    4. Discuss other possible apporaches for achieve additional compression ratio. 
  4. Using Multimedia in Web Page
    1. How big is a raw 2 minutes, mono, audio file sampled with 44kHz and 16 bits per sample?
        2.  
        1. 352000 bytes. Yes No
        2. 21120000 bytes. Yes No
        3. 2640000 bytes. Yes No
    2. How to reference video?
      3. 1. Use <A HREF="video.avi"> Yes No
      2. Use <Object data="video.avi"> Yes No
      3. The  <Object data="video.avi">displays the video image on the same web page. Yes No
      4. The <A HREF="video.avi"> embeds the video image on the same web page. Yes No